Legacy Assembler Transformation FAQ
The following are some frequently asked questions regarding SML’s approach to legacy Assembler systems:
Complicated systems, often the result of M&As, are a barrier to the level of transparency required by regulatory bodies, especially in the banking sector. Whilst other programming languages can be more easily upgraded, Assembler stubbornly resists comprehension, except by the very few (who are rapidly retiring).(Up)
SML provide the migration as a fixed price service. It is charged on a price per line of code plus consultancy required to customise the migration engine for customer specifics. For further information contact SML directly.(Up)
Converting your legacy system to a newer language means that the system is better understood by a larger team of people. It also dramatically reduces the maintenance and development costs of adapting Assembler for the 21st century. SML estimates that Assembler costs 4 times more to support and 6 times more to develop in than a new platform.(Up)
The SML approach to migration can save in excess of 80% compared to a manual conversion process.(Up)
Companies can increase team productivity, system longevity and sustainability through SML’s Workbench. This solution improves system knowledge which can produce up to 30% reduction in maintenance effort and time to market. The Workbench provides a return on investment in less than 6 months (usually less than 4).(Up)
Once the parsers are tuned for the customer, the automated migration process takes very little time, for example 500,000 lines of assembler were recently converted on a single 2.6Ghz processor in 90 minutes. This means that even the largest systems can be migrated in a matter of hours by performing the migration process in parallel on several different machines.
This ability to migrate en masse very quickly has a significant impact on the migration as a whole. Using normal migration techniques it would be necessary to "freeze" the code that is to be migrated, or use a very complex change management system until the migration is complete. Using the FermaT Migration Engine it is possible to continue performing system changes and enhancements right up to the point where the whole system is to be migrated. It is then "frozen" whilst the automated migration takes place and the code is returned to the customer for immediate regression and system testing. Indeed if required the original system can continue to be changed and enhanced whilst the regression testing takes place. Once this has been completed we can rerun the automated migration process.
This has significant business benefits, customers can continue to meet urgent business requirements whilst the migration is taking place ensuring they do not miss out on opportunities or competitive advantage. (Up)
Many CIOs may not know how much Assembler they have in their system, they just know that it takes 18 months to get a new product to market, whilst the business is demanding it take 3 months. The complexity of adapting Assembler, or even knowing what it is doing, is often the cause of the delay. SML can get your core systems working for your company rather than holding it back.
SML’s automated, iterative and demonstrable process also mitigates exposure during the migration.(Up)
CIOs have spent the last few years reducing costs and managing outsourcing contracts. SML believes that 2016 will be the year for corporate migrations as CIOs realise the benefits of getting rid of costly and complex core systems. Migration is no longer a cumbersome, risky process; SML has unique product offerings of system comprehension and migration, which will analyse, document, re-engineer and migrate complex code via its proprietary technology, generating 100% of the transformed code automatically.
Companies can improve system productivity and reduce the costs of maintaining and working with Assembler.(Up)
Migrating to a new platform has now become a real option for the cutting-edge CIO. It will avoid the risk of an aging system failing at an inopportune moment and the retirement of the only guy who knows how the system truly works! SML is the crucial piece of the migration jigsaw; shining a light into the dark depths of Assembler code and converting it into something useable, in a measurable, iterative process, without having to shut down the system or rebuild it from scratch.(Up)
Assembler is a very flexible low level programming language. When most of the Assembler applications were being written system resources were at a premium and developers were able to use a whole host of techniques such as complex register manipulation, use of off-sets, etc. Many Assembler programs lack any useful documentation and have become over complicated following years of maintenance.
With the exception of the FermaT Analysis Workbench there are very few specialist Assembler Analysis tools that enable the developer to quickly and easily understand and document an Assembler program’s functionality. The inherent complexity of most Assembler code combined with a lack of tools and a diminishing pool of skilled Assembler resources makes doing anything with Assembler much more difficult and costly than other mainframe development languages such as COBOL.
When it comes to transformation, the problems lie in ensuring the quality and maintainability of the transformed code. Some Assembler migration solutions on the market today merely take complex, badly structured code and replicate it in another language such as C. This is not an effective solution as it merely transposes the legacy problem from one language to another.
Because of FermaT's use of a formally defined intermediate language and powerful code transformations the C or COBOL code generated from the Assembler is easier to maintain and of a consistently high quality. (Up)